HAL Id: hal-02397302
https://hal-univ-rennes1.archives-ouvertes.fr/hal-02397302
Submitted on 24 Mar 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
New generations of spirobifluorene regioisomers for
organic electronics tuning electronic properties with the
substitution pattern
Cyril Poriel, Lambert Sicard, Joëlle Rault-Berthelot
To cite this version:
Cyril Poriel, Lambert Sicard, Joëlle Rault-Berthelot. New generations of spirobifluorene
regioiso-mers for organic electronics tuning electronic properties with the substitution pattern. Chemical
Communications, Royal Society of Chemistry, 2019, 55 (95), pp.14238-14254. �10.1039/c9cc07169e�.
�hal-02397302�
a. Univ Rennes, CNRS, ISCR-UMR 6226, F-35000 Rennes, France.
New Generations of Spirobifluorene Regioisomers for Organic Electronics: Tuning
Electronic Properties with the Substitution Pattern
Cyril Poriel,a* Lambert Sicard,a Joëlle Rault-Berthelota
The spirobifluorene (SBF) fragment constitutes one of the most important scaffold used in the design of Organic Semi-Conductors (OSCs) for organic electronics. For the last ten years, new generations of SBF positional isomers have appeared in the literature. The different positions of substitution (C1, C3 or C4) have allowed the tuning of the electronic properties of great interest for the further design of functional materials. The high potential of these new generations of organic semi-conductors in electronics has been demonstrated notably when used as host for Phosphorescent Organic Light-Emitting Diodes (OLEDs) or for Thermally Activated Delayed Fluorescence OLEDs. In the present feature article, we present these new generations of SBF compounds and the impact of positional isomerism on the electronic properties and devices performance. Particularly, we show how the different structural and electronic parameters (nature of the linkages, bridge substitution and steric hindrance) drive the electrochemical and photophysical properties of SBF regioisomers and can be modulated. Such studies lay the foundations of materials design for organic electronics.
Introduction
Spiro configured compounds constitute one of the most important class of Organic Semi-Conductors (OSCs) for electronics.1-3 Since the display of the 'spiro concept' in the
nineties by the group of Salbeck, the 9,9’-spirobifluorene (SBF) has become a central molecular scaffold in organic electronics.3, 4 SBF is the association of two fluorene units via a shared spiro carbon (Figure 1-Left). It possesses a particular 3D geometry with the two fluorene units set along two orthogonal planes. One of the particularity of the SBF fragment is its capacity to improve the thermal and morphological properties of the OSC in which it has been introduced.2
Therefore, the SBF scaffold is found in many highly efficient OSCs especially for Organic Light-Emitting Diodes (OLED) as a fluorophore5 or as high triplet host material for phosphors.6 In
the field of solar cells, the SBF fragment also played a key role as the widely known 2,2',7,7'-Tetrakis[N,N-di(4-methoxyphenyl)amino]-9,9'-spirobifluorene (Spiro-OMeTAD), used as hole transporting material, is constructed on a SBF core.7-9 New electronic applications taken advantages of the
cross-shape geometry of the SBF fragment have also started to appear in the literature such as non-fullerene acceptors in solar cells,10-13 spiro-fused nanographene structures14 or
ordered monolayers.15, 16
Thanks to this singular geometry, the SBF scaffold has also been investigated for other appealing applications outside of organic electronics such as fluorescent marker for biomolecules,17 chiral ligands,18, 19 catalysts in homogeneous20,
21 or heterogeneous21-24 chemical reactions (epoxidation,
sulfoxidation…), or as building units in coordination polymers,25, 26 showing the versatility of this platform. In all
these examples, it is the 3D geometry of the SBF scaffold that has been taken advantage of. The substitution pattern of SBF has been far less studied and it is the main purpose of the present feature article.
In the SBF fragment, sixteen substitution positions are available, 4 on each phenyl unit numbered from 1 (in β position of the spiro carbon) to 4 (in α position of the biphenyl linkage), Figure 1-Left. 2-Substituted SBFs were the first generation to be developed,2 due to an easier synthetic access
compared to the other positional isomers. Indeed, the direct aromatic electrophilic substitution of the SBF core takes place at C2 and this type of reaction has been a precious tool to develop 2-substituted SBFs. The para linkage between the pendant substituent at C2 and the constitutive phenyl rings of the fluorene ensures a good delocalization of π-electrons, essential to create efficient fluorophores.2-5 However, in recent
years, the growing necessity to design efficient host materials for blue Phosphorescent Organic Light-Emitting Diodes (PhOLED)27, 28 has led to a demand of new generations of
SBF-based materials with wide energy gaps (ca 4 eV) and hence a restricted π-conjugation. Indeed, in order to obtain a high triplet energy (ET), a key feature in the design of host materials
for blue PhOLEDs (which are still the weakest link of this technology),29 the π-electrons delocalization within the OSC
has to be restricted. This π−conjugation disruption has been successfully investigated with ortho linked SBFs (substitution at position C4,3, 30-35 5 first example reported in 2009) and meta
linked SBFs (substitution at position C3-first example reported in 2013 or at position C1-first example reported in 2017),36-39
leading to high-efficiency green and blue PhOLEDs. However and despite these recent high-performing devices, only few examples of 1-, 3- and 4-substituted SBFs have been described to date. Nevertheless, they possess great potential.
Figure 1. Nomenclature of SBF (Left) and schematic representation of the SBF substitution positions (Right).
In May 2017, the first structure-property relationship study covering the four positional isomers of SBF was reported by
9 1 2 3 4
Accepted
Manuscript
our group.39 This work has not only highlighted the strong
effect of the substitution pattern on the electronic properties but has also revealed the potential of the position C1 for electronic applications. In March 2019, the efficiency of the C1 position to construct high ET SBF-based host materials was
demonstrated, leading to the highest performance ever reported for pure hydrocarbon hosts in blue PhOLEDs.6 This
important finding has motivated the present review. Herein, we aim to overview these new generations of SBF isomers substituted at C1, C3 or C4 and the effect of the substitution pattern in modulating their electronic properties. As regioisomerism is an important concept, which drives the properties of OSCs and their resulting device performance,1, 39-43 the goal of this work is to define the general rules of SBF
regioisomerism in order to help the rational design of future functional materials for electronics.
In the present article, we will first focus on the general synthetic routes, which have been developed over the last years to access these new generations of SBF regioisomers. Then, through a structure-property relationship approach (exemplified with the four phenyl SBF isomers), we will describe the impact of regioisomerism on the electronic properties (electrochemistry, absorption, fluorescence, phosphorescence). Finally, selected examples of OSCs belonging to each family of SBF (substituted either at C4, at C3 or at C1) will be discussed. The potential of these OSCs as host materials for PhOLEDs or for Thermally Activated Delayed Fluorescence (TADF) OLEDs will also be highlighted. Thus, in the present work, we discuss how the different structural and electronic parameters (nature of the linkages, bridge substitution and steric hindrance) drive the electronic properties of SBF regioisomers. Such studies lay the foundations of materials design for organic electronics.
In order to well describe the impact of the substitution of the SBF core, this review will only focus on monosubstituted SBFs.
Part 1. Synthetic Investigations
To date, SBF compounds substituted at C2 have been far more developed than those substituted at the three other positions. A review was published by Salbeck and coworkers in 2007.2
Indeed, incorporation of a molecular fragment at positions C1, C3 or C4 of a SBF backbone is far more complicated than at position C2 since the direct electrophilic substitution of SBF
does regioselectively occur on the latter.2 This fact has hindered the development of C1, C3 or C4 positional isomers. However, the last ten years have led to the development of efficient synthetic routes towards these regioisomers, which are at the origin of the rise of these materials in organic electronics. In this context, the halogeno derivatives hold an important place and the 1-, 2-, 3- and 4-bromo-9,9’-spirobifluorenes (1-Br-SBF, 2-Br-SBF, 3-Br-SBF and 4-Br-SBF)
are the cornerstone of all the substituted SBFs reported to date.
First reported in 1930,44SBF is classically obtained in a
two-step synthesis from the coupling of 2-halogenobiphenyl (1 or 2) and 9-fluorenone 3 followed by an intramolecular
cyclization of the resulting fluorenol to form the spiro bridge. The synthesis of 1-Br-SBF, 3-Br-SBF and 4-Br-SBF platforms is
based on the introduction of the bromine atom before this key cyclization step. Theoretically, the introduction of the bromine
atom can be done either on the electrophile (fluorenone) or on the nucleophile (biphenyl). However, incorporating the bromine atom on the nucleophile is difficult and has only been performed in the case of 4-Br-SBF (see below).
Thus, the approach towards bromo-spirobifluorenes (1-Br-SBF, 2-Br-SBF, 3-Br-SBF and 4-Br-SBF (Scheme 1) consists in fixing
the bromine atom on the fluorenone core prior to the final cyclization. This route involves the synthesis of the
corresponding fluorenones substituted either at C1 (1-Br-FO),
C2 (2-Br-FO), C3 (3-Br-FO) or C4 (4-Br-FO). As 2-Br-SBF and its
corresponding fluorenone 2-Br-FO have been largely reported
in the literature,2 they will not be described herein and we will only focus on the three other regioisomers.
Scheme 1. Retrosynthetic analyses of bromo-spirobifluorenes 1.a. Synthesis of 4-Br-SBF and 4-substituted SBFs
Firstly, in the case of 4-Br-SBF, the fluorenone 4-Br-FO was reported
by a selective Miyaura-Suzuki cross coupling between 2-ethyl carboxylate phenylboronic acid 5 and 2-bromoiodobenzene 4
followed by an intramolecular aromatic electrophilic cyclization of the resulting biphenyl 6 (Scheme 2).32 The anchoring of a
spiro-connected fluorene unit on 4-Br-FO using bromobiphenyl 1 leads to
the fluorenol 7, which finally provides 4-Br-SBF after an
intramolecular ring closure step.
Br 4-Br-SBF HO Br AcOH/HCl 3 83% (2 steps) O 8' Br O CO2Et Br 6 4-Br-FO MsOH, 100°C (95 %) AcOH/HCl 70°C (75 % from 4-Br-FO) Br HO n-BuLi/Ph2Br1 THF -78°C to rt I Br CO2Et B(OH)2 + Li Br n-BuLi/THF -78 °C Br Br 1) CO2 -78°C to rt 2) H2O 3) H2SO450°C 8 (42 % from 8) (65 %)K2CO3, Pd(dppf)Cl2 Toluene, 100 °C 9 4 5 7
Scheme 2. Synthesis of 4-Br-SBF and 4-Br-FO
In 2009, the group of Ma also reported the synthesis of the key building block 4-Br-FO thanks to a mono lithium-halogen exchange
of 2,2'-dibromobiphenyl 8 followed by the trapping of the
corresponding lithiated intermediate 8’ with carbon dioxide
(Scheme 2).31 Secondly and as mentioned above, the bromine can also be attached to the nucleophile (ie the biphenyl). Thus, the trapping of the mono-lithiated intermediate 8’ can be done with 3
4-Br-SBF 3-Br-SBF 2-Br-SBF 1-Br-SBF X O HO 4-Br-FO 3-Br-FO 2-Br-FO 1-Br-FO Br Br Br X= Br X= I 1 2
Accepted Manuscript
directly providing 4-Br-SBF after cyclization of the fluorenol
intermediate 9.31 Most of the 4-substituted SBFs reported in
literature were synthesized from Br-SBF or its pinacol analogue
4-pinacolborane-SBF. They will not all be reported herein as they are covered by an exhaustive review published in 2017.3 They
incorporate many different functional units such as electron- or hole-transporting fragments (Scheme 3), which allow to adjust molecular orbitals and to use 4-substituted SBFs as functional materials in electronic devices (see Part 3).
4-Ph-SBF 4-Br-SBF K2CO3,Pd(dppf)Cl2 DMF, C6H5B(OH)2 150°C (91%) 4,4'-(SBF)2 K2CO3,Pd(PPh3)4, THF 4-Pinacolborane-SBF Reflux (72%) X X B(OH)2 THF/H2O, reflux X= O (75%) X= S (80 %) 4-DBF-SBF K2CO3Pd(PPh3)4 K2CO3Pd(dppf)Cl2 DMF, 150°C 4-PhCbz-SBF N N B (75%) (83%) OMe MeO MeO 4-PhOMe3-SBF K2CO3Pd(dppf)Cl2 DMF, 150°C OMe OMe OMe (HO)2B 4-POPh2-SBF 1/ n-BuLi/ClPPh2 P Ph Ph O 2/ H2O2 (18%) 4-4Py-SBF 4-3Py-SBF N B(OH)2 K2CO3Pd(dppf)Cl2 DMF, 150°C N N B(OH)2 (89 %) (86 %) 4-5Pm-SBF (94%) N N O O B(OH)2 Toluene, 110°C 4-DBT-SBF OC16H33 C16H33O C16H33O (48%) 4-EPHDB-SBF C16H33O C16H33O OC 16H33 n-propylamine Pd(PPh3)4 K2CO3, Pd(PPh3)4 OC16H33 OC16H33 OC16H33 O H N 4-EPHDBA-SBF N H O OC16H33 OC16H33 OC16H33 (51%) n-propylamine Pd(PPh3)4 N N BO O N N (51%) Br 4-PPI-SBF
Scheme 3. Synthesis of selected examples of 4-substituted SBFs from 4-Br-SBF 1.b. Synthesis of 3-Br-SBF and 3-substituted SBFs
3-substituted-SBFs have been less developed than their 4-substituted isomers described above due to synthetic difficulties. In principle, the general routes described above can also be used to synthesize 3-Br-SBF (incorporation of the bromine atom either on
the fluorenone or on the biphenyl). However, as far as we know, only the first approach has been used so far. Thus, 3-bromo-9-fluorenone 3-Br-FO appears as a key intermediate in the synthesis
of 3-Br-SBF (Scheme 4) and different routes have been reported.
Liao, Jiang and coworkers have used an elegant Pschorr cyclization reaction to form the fluorenone backbone from benzophenone 10
after a diazotation step (55% yield), Scheme 4-Top.38, 45 This reaction is regioselective providing only 3-Br-FO.
O NH2 Br 1) H2O/H2SO4, 60°C 2) NaNO2, 0°C - 50°C O Br 3-Br-FO 55% HO n-BuLi/Ph2Br 1 THF -78°C to rt Br 3-Br-SBF AcOH/HCl reflux (67% - 3 steps) Br O O Br O O Br2, hn 1. KOH 2. KMnO4 55% 52% 10 11 12 13 benzoyl peroxide
Scheme 4. Synthetic approaches towards 3-Br-SBF via 3-Br-FO
The notion of regioselectivity vs non-regioselectivity is an important point in our discussion and a modern concern in the field of OSCs for organic electronics.1, 46, 47 If in principle, this approach could be used to synthesize the other isomers of bromofluorenone by modifying the position of the bromine atom in the starting material
10, it will most likely suffer from a selectivity problem in the case of 2-Br-FO and 4-Br-FO. It should be noted that, in theory, 3-Br-FO can
also be obtained following a non-regioselective intramolecular cyclization of the biphenyl 24 (see Scheme 6).
Phenantroquinone 11 has also been used in the synthesis of 3-Br-FO following a bromination /oxidation sequence. Thus, 11 is first
brominated to provide 12 before releasing 3-Br-FO by the cleavage
of the C/C bond linking the two carbonyl groups under the action of KOH and KMnO4 (29 % over the two steps, Scheme 4, Bottom).48
From 3-Br-FO, 3-Br-SBF was then synthesized following the classic
sequence of spirofluorene introduction (formation of fluorenol 13
and cyclization). All the 3-substituted SBFs reported in the literature (except 3-Ph-SBF) were synthesized from 3-Br-SBF platform,
Scheme 5. Notably, electron withdrawing (such as phosphine oxide in 3-POPh2-SBF)45 and/or electron donating units (such as carbazole
oligomers in 3-diNCbz-SBF38) were introduced providing efficient
functional materials as described in Part 4.
3,4''-(SBF)2 3-Br-SBF THF, Reflux (86 % for 3,4''-(SBF)2) (74% for 3,3''-(SBF)2) Pd(PPh3)4, K2CO3 THF, 60°C K2CO3, Pd(PPh3)4 3-POPh2-SBF 1/ n-BuLi/ ClPPh2 2/ H2O2 (54%, two steps) PO P N N N O N O O B N N B THF, 60°C K2CO3, Pd(PPh3)4 (95%) N HN DMF, 150°C CuI, Phenanthroline K2CO3 1/ n-BuLi/ Cl2PPh 2/ H2O2 (42%, two steps) O O 3-POPh-(SBF)2 3-(N-Ph-Cbz)-SBF 3-diNCbz-SBF 3-(3-PhCbz)-SBF (92%) (83%) 3,3''-(SBF)2 Br 3-B(OH)2-SBF or 4-B(OH)2-SBF
Scheme 5. Synthesis of 3-substituted SBFs via 3-Br-SBF
Accepted
1.c. Synthesis of 1-Br-SBF and 1-substituted SBFs
1-substituted SBFs are far less developed than the two other families described above and represent the youngest generation of SBF positional isomers. To the best of our knowledge, the first example of a 1-substituted SBF (1-Ph-SBF) for organic electronics
was synthesized in May 2017 from 1-iodofluorenone 1-I-FO39,
although some examples had been reported earlier.49 As for the
two preceding families of positional isomers, 1-halogenofluorenones (1-I-FO and 1-Br-FO) are key compounds in
the synthesis of 1-substituted SBFs.
Synthetic investigations towards such 1-halogenofluorenones have recently encountered a significant development, which will clearly help building 1-substituted SBFs in the future.
In 1951, 1-aminofluorenone 17 was reported by Kharash and Bruice
from the oxidation of fluoranthene 14 in 1-carboxylic acid
fluorenone 15 (Scheme 6)50 allowing large scale synthesis of this compound (note that an earlier work reported amine 1751). This
strategy is appealing as it allows in one step to build the fluorenone core with a substituent, ie a carboxylic acid, located at C1. The carboxylic acid 15 is then converted to its amide 16 and further
converted to its amine 17. In 2016, the group of Harper52 screened
various synthetic pathways towards 1-substituted fluorenones. This group notably revisited the original synthesis of 1-I-FO and
delivered many different synthetic pathways towards other 1-substituted fluorenones such as 1-Br-FO. For example, a
regioselective approach towards 1-Br-FO was explored. The
approach is similar to that presented above for 4-Br-SBF (Scheme
2). However, to reach 1-Br-SBF, the bromine atom must be located
at the α position of the carboxylate group on biphenyl 21 (whereas
the bromine atom and ethyl ester group were each located on a different phenyl ring in 6, Scheme 2). Thus, after a selective iodation
of 18 providing 19 and protection of the carboxylic acid as a methyl
ester group (20), biphenyl 21 is involved in an intramolecular
cyclization regioselectively providing 1-Br-FO.
In 2016, Bentabed-Ababsa et al reported that the direct iodation at C1 of fluorenone 3 in a 52% yield was possible when using an in situ
deprotolithiation-zincation sequence.53 Other side products were nevertheless detected during this reaction.
NH2 O 1) H2O/HBr, 0°C 2) HBr, CuBr 40% NH2 CO2Et MsOH 130°C-43% piperidine NaOH 24 26 17 CO2H Br 18 I2, PhIOAc, Pd(OAc)2 DMSO CO2R Br CO2Me 21 I Br 1/ NaOH 2/ SoCl2 3/ AlCl3 B(OH)2 76% DBU/MeI 100°C 90% NH2 25 MsOH 130°C H O Br I + 10 mol % Pd(OAc)2 2.0 eq AgTFA AcOH/HFIP (9:1, 0.15M) 120°C, air, 38 % NH2 CO2H 40 mol % CO2H O 15 CrO3/ AcOH 80/100°C CONH2 O 16 1) SOCl2 2) NH3aq 76% 1) NBS/DBU MeOH 87% 2) TBAF/THF 60% 8 % 1-Br-FO O Br 1) H2O/HCl, 0°C 2) NaNO2/H2O, 0°C 3) KI/H2O 75 % 3 1) ZnCl2·TMEDA THF 2) BuLi / 2,2,6,6-tetramethylpiperidine 3) I2 52% 14 28 29 N O H H
Intramolecular O-H interaction occuring in 1-aminofluorenone 17 3Br-FO O 1-I-FO O I O Br O 25% 99% NH2 CN CN O H CH2(CN)2 CO2Et B(OH)2 Pd(dppf)Cl2 K2CO3 DMF, 130°C 23 41% 27 Pd(OAc)2, Sphos Na2CO3, 94% PPA, 100°C 55 % R=H (19) R=Me (20) 2-bromoaniline 22
Scheme 6. Synthesis of 1-substituted SBFs 1-Cbz-SBF and 1-Ph-SBF
The approach developed by our group towards I-FO via the
1-aminofluorenone 17 presented the advantage to also lead to
another positional isomer, ie 3-aminofluorenone 25, of interest in
the purpose of this article.39 This route was based on a Pd
cross-coupling between 3-bromoaniline 22 and (2-(ethoxycarbonyl)phenyl)boronic acid 23 to give the biphenyl 24
further cyclized in methanesulfonic acid to provide 17 and 25 in 43
and 25 % yield resp.. Thanks to the rotation of the C/C bond of the biphenyl linkage in 24, the cyclization can occur on the two ortho
positions of the aminophenyl core. This cyclization step is therefore not regioselective. Fortunately, the authors reported that the separation of the two aminofluorenones 17 and 25 on column
chromatography was easy due to the different interactions with silica gel. Indeed, in 17, the proximity between the oxygen atom of
the ketone and the amino group leads to intramolecular hydrogen bond (Scheme 6, inset), which is not the case with isomer 25. As it is
known that the ratio of regioisomers formed in such a type of cyclization can be tuned by varying experimental conditions such as the temperature of the reaction or the solvent,1, 46, 47 this approach
provides a promising synthetic route. In other hand, a regioselective strategy towards 17 was developed by Velasco and Yu.54 They have
reported that 3-aminobiphenyl-2,4-dicarbonitrile 27 in
polyphosphoric acid can undergo both a cyclization and decyanation step providing 1-aminofluorenone 17 (55% yield). The
substitution of the 17 by an iodine atom was then classically
performed via a Sandmeyer reaction yielding 1-I-FO with 75% yield.
39
Finally, the Sorensen’s group has reported an elegant and efficient one-pot synthesis of 1-Br-FO via a Pd(II)-catalysed C(sp2)
functionalization cascade starting from iodobenzene 28 and
2-bromo-benzaldehyde 29 and using anthranilic acid as transient
directing group55 This approach is rapid, efficient (38% yield) and, in
our point of view, the best synthetic strategy reported to date. Thus, in the last three years, several synthetic investigations towards 1-halogenofluorenones were carried out. These studies provide different synthetic routes to these key molecules, which are indeed of great interest to construct functional materials.
Nevertheless, the literature does not yet report any 1-halogeno-SBF. Indeed, in the 1-substituted SBFs reported to date, the substituent was introduced on the fluorenone backbone prior to the spirolinked fluorene unit (for example from fluorenone 30 for 1-Cbz-SBF or from fluorenone 31 for 1-Ph-SBF, Scheme 7). We believe
that the successful synthesis of 1-halogeno SBFs will be an important step to increase the diversity of 1-substituted SBF based compounds. CuO, TMEDA K3PO4, Carbazole 1,2-dichlorobenzene 180°C 95% Br O Pd(dba)2, PCy3, KF PhB(OH)2 DMF, 130°C 85% O N 1) 2-iodobiphenyl 2 / nBuLi THF, -78°C 2) AcOH/HCl reflux 68% N 30 1-Cbz-SBF I O 1-Br-FO 1-I-FO O 31 1-Ph-SBF 1) 2-iodobiphenyl 2/ nBuLi THF, -78°C 2) AcOH/HCl reflux 38%
Scheme 7. Synthesis of 1-substituted SBFs 1-Cbz-SBF and 1-Ph-SBF
In March 2019, C1-linked SBF dimers were reported (Scheme 8).6 These molecules have shown the highest performance reported to date for a Pure HydroCarbon (PHC) material as host in blue
Accepted
PhOLEDs (see part 5). This shows the strong potential of the C1-SBF scaffold.
The four C1-linked SBFs, 1,2’’-(SBF)2, 1,3’’-(SBF)2 and 1,4’’-(SBF)2
were synthesized following a similar approach starting from their corresponding pinacol derivatives, 2-Bpin-SBF, 3-Bpin-SBF and 4-Bpin-SBF (Scheme 8, top). These pinacols were first coupled with
1-bromofluorenone 1-Br-FO to provide the corresponding
fluorenones 1,2’-FO-SBF, 1,3’-FO-SBF and 1,4’-FO-SBF with high
yields (70 to 85%) showing that these reactions were weakly dependent of the SBF substitution pattern. Classically, these fluorenones were converted to their corresponding dimers with high yields. Again, the authors noted that despite a strong sterically hindered environment, the spirolinked fluorenes can be efficiently introduced in the last step with the pending SBF already in place. This feature should be advantageously used in the future to construct C1-linked SBF materials. However, due to this steric congestion, 1,1’’-(SBF)2 was reported through a different synthetic
pathway (Scheme 8, Bottom). 1-Br-FO was first dimerized through a
one-pot Pd-catalysed coupling to give the difluorenone dimer 1,1’-(FO)2 before forming the bis-diol intermediate 32 further cyclized in
1,1’’-(SBF)2 with HBr/AcOH. Pd(dppf)Cl2.2 CH2Cl2 K2CO3 DMF, 150°C, 15h O O B O OH 2-Bpin-SBF 3-Bpin-SBF 4-Bpin-SBF 1-2'-FO-SBF (85%) 1-3'-FO-SBF (70%) 1-4'-FO-SBF (73%) 1-2''-(SBF)2 (78%) 1-3''-(SBF)2 (90%) 1-4''-(SBF)2 (68%) 2) RT, 24 h HCl / AcOH reflux Bpin 2 3 4 Bpin = 1-Br-FO 2 3 4 1) Ph2I 2 n-BuLi, THF, -78°C 2 3 4 2 3 4 3) EtOH O O (78%) Ph2Br 1 /n-BuLi THF / -78°C 1) B-pin, Pd(dppf)Cl2, KOAc Dioxane, 100°C, 12 h 2) 1-Br-FO, Pd(PPh3)4, K2CO3 Dioxane, 100°C, 12 h 1-1'-(FO)2 OH HO 1-1''-(SBF)2 HBr/AcOH reflux (30%, two steps) 32 O Br 1-Br-FO O Br
Scheme 8. Synthesis of the C1-linked SBF dimers: 1,2’’-(SBF)2, 1,3’’-(SBF)2 and 1,4’’-(SBF)2 (Top) and 1,1’’-(SBF)2 (Bottom)
To conclude, efficient synthetic routes have now been developed for all the SBF regioisomers. This was a key point and the first step towards their use in materials science and electronics
.
Part 2. Influence of the SBF substitution pattern
on the electronic properties: Focus on the origin
of the π-conjugation breaking
In order to precisely highlight the differences linked to the substitution pattern related to the electronic properties, this
second part will focus on the only work described to date that covers the four SBF isomers. These four isomers, 1-Ph-SBF, 2-Ph-SBF, 3-Ph-SBF and 4-Ph-2-Ph-SBF, are all substituted with a simple
phenyl ring (Figure 2- Top).39 The original question of this pioneer work was: What is the impact of the substitution on the electronic
properties and particularly on the ET, key property for PhOLED
applications? Two important parameters, driving these properties,
will be discussed in detail: the electronic parameter (nature of the linkage and bridge substitution) and the steric parameter (fluorene/fluorene dihedral angle). The electronic properties are gathered in Table 1.
The chief structural parameter driving the electronic properties is the relative position of the pendant substituent (herein the phenyl ring) with respect to the fluorene.34, 35 Thus, 2-Ph-SBF possesses a
fluorene/phenyl dihedral angle of 37.4°, very similar to that of SBF, 34.2° (values obtained from X-ray, Figure 2-Top). In 3-Ph-SBF, the meta linkage between the pendant phenyl ring and its
resulting electronic decoupling should strongly reduce the conjugation between the two fragments. We will see later that it is not as simple.
Table 1. Electronic data of 1-Ph-SBF, 2-Ph-SBF, 3-Ph-SBF, 4-Ph-SBF and SBF.
a. in cyclohexane, b. from CVs, c. from UV-Vis spectra, d. in 2-Me-THF at 77K.
1-Ph-SBF 2-Ph-SBF 3-Ph-SBF 4-Ph-SBF SBF λabs (ε)a (nm) (×104 L mol-1 cm-1) 298 (0.98) 309 (1.66) 296 (2.44) 308 (2.31) 319 (1.60) 297 (0.92) 310 (1.49) 316 (0.64) 297 (1.07) 309 (1.49) 297 (0.72) 308 (1.45) λema (nm) 313, 323 334, 350 332, 343 359 310, 323 QYa 0.61 0.87 0.74 0.42 0.40 τfluo (ns)a 5.16 1.56 5.74 4.20 4.60 kr (×108) (s-1) 1.22 5.60 1.29 1.00 0.87 knr (×108) (s-1) 0.72 0.83 0.45 1.40 1.30 HOMOb (eV) -5.94 -5.86 -5.94 -5.95 -5.95 LUMOb (eV) -1.73 -1.99 -1.77 -1.87 -1.74 ΔE (eV) Opt c 3.95 3.70 3.78 3.82 3.97 Elb 4.21 3.87 4.17 4.08 4.21 ET (eV)d 2.86 2.56 2.83 2.78 2.88 Τphospho (s)d 5.8 3.3 5.4 4.7 5.3 1-Ph-SBF 3-Ph-SBF 4-Ph-SBF 2-Ph-SBF
Accepted
Manuscript
Metalinkage
4-Ph-F 2-Ph-F
3-Ph-F 1-Ph-F
Meta linkage Para linkage
Ortholinkage F 1 2 3 4 290 300 310 320 330 340 350 0 5000 10000 15000 20000 25000 E ps il on (m ol .L -1.c m -1) Wavelength (nm) SBF 1-Ph-SBF 2-Ph-SBF 3-Ph-SBF 4-Ph-SBF 300 350 400 450 500 0.0 0.2 0.4 0.6 0.8 1.0 N orm al iz ed P L (a . u.) Wavelength (nm)
Figure 2. 1-Ph-SBF, 2-Ph-SBF, 3-Ph-SBF and 4-Ph-SBF. Top. ORTEP drawing (ellipsoid probability at 50 % level) from X-Ray crystallography, Bottom. Absorption (Left) and emission (Right) spectra in cyclohexane (SBF is added for comparison purpose)
In 1-Ph-SBF and 4-Ph-SBF, the phenyl/fluorene dihedral angle
is impressively larger, 75.4° and 51.2° respectively. This is due to a higher steric hindrance between the pendant phenyl ring and either the cofacial fluorene in the case of 1-Ph-SBF39 or
the hydrogen atoms of the substituted fluorene in the case of
4-Ph-SBF.35 Therefore, two molecules possess a small angle
(2-Ph-SBF and 3-Ph-SBF) and the two others a large one
(1-Ph-SBF and 4-Ph-(1-Ph-SBF). This structural characteristic will be one of
the key parameters involved in the different electronic properties described below for all the SBF isomers. Let's first have a look on the consequences on the absorption properties. First, we need to remind that unsubstituted SBF exhibits two
characteristic absorption bands at 297 and 308 nm (π−π∗ transitions).35 The four phenyl-substituted SBF isomers all
display these two bands (Figure 2-Bottom, left). In addition to these bands, 2-Ph-SBF displays a large one at 319 nm,
translating an extension of the conjugation to the pending phenyl unit. This extension of conjugation was assigned to the combination of two parameters: the para linkage (positional effect) and the small dihedral angle (steric effect) adopted between the pending phenyl and the fluorene. Instead of this large band at 319 nm, the spectrum of 4-Ph-SBF presents a
weak band tail between 309 and 325 nm signing a conjugation disruption due to the large angle formed between the fluorene and the phenyl at C4.35 In this case, the steric effect is
predominant over the electronic effect as an ortho linkage should in principle allow a similar electronic coupling than a
para one.
The meta linkage of both 1-Ph-SBF and 3-Ph-SBF has revealed
different behaviours. Indeed, the absorption spectrum of 3-Ph-SBF displays a large band at 316 nm, very similar to that of 2-Ph-SBF with nevertheless a molar absorption coefficient 2.5
times lower (Table 1). This extension of the π-conjugation appeared surprising in the light of the literature as it was commonly admitted that there is a better delocalisation of π-electrons following the para/ortho/meta sequence.48, 56-61
As 3-Ph-SBF presents a relatively intense degree of
conjugation between the phenyl and the fluorene, its behaviour is different to that of its building block meta terphenyl (in absorption spectroscopy, meta terphenyl displays a maximum at 246 nm and para terphenyl at 277 nm).62 The
conclusion drawn by the authors was that the 'linkage' effect could not totally explain this feature and other parameters should be invoked. The authors have assigned this particularity to the presence of the spiro bridge.
290 300 310 320 330 340 350 0 5x103 1x104 2x104 2x104 3x104 3x104 4x104 E ps il on (m ol .L -1.c m -1) Wavelength (nm) F 1-Ph-F 2-Ph-F 3-Ph-F 4-Ph-F
Figure 3. Absorption spectra in cyclohexane (Left) and molecular structures (Right) of 1-Ph-F, 2-Ph-F, 3-Ph-F and 4-Ph-F and F.
By studying model compounds with different bridge substitution (the four isomers of fluorenes 1-Ph-F, 2-Ph-F, 3-Ph-F and 4-3-Ph-F, Figure 3-Right), the authors have shown that
this bridge substitution was strongly involved in the extension of the conjugation observed in 3-substituted SBFs. Indeed, both Ph-SBF (Figure 2- Bottom, left) and Ph-F (Figure
3-Left) display a large band characteristic of the electronic coupling between fluorene and phenyl but possess interestingly different intensities. It is clear that this bridge effect will deserve to be more investigated in the future in order to control and tune the photophysical properties of the resulting materials.
The other meta-linked SBF, 1-Ph-SBF, displays a complete
π-conjugation breaking (Figure 2-Bottom, left) as its reported UV-vis absorption spectrum is almost identical to that of SBF.
This breaking arises from two parameters: the meta linkage (which cannot completely break the conjugation as exposed above for 3-Ph-SBF) and the very large phenyl/fluorene
dihedral angle caused by the presence of the cofacial fluorene. This has been confirmed by the absorption spectrum of the corresponding model compound 1-Ph-F, which is very different
as it displays a long tail (Figure 3-Left), reflecting a certain degree of conjugation between the pending phenyl and the fluorene moiety. Indeed, the pendant phenyl ring in 1-Ph-F is
not sterically hindered oppositely to that of 1-Ph-SBF. Thus,
the nature of the linkage (electronic effect) and its position (steric effect) are two key parameters of the π-electrons delocalization. Nevertheless, the bridge substitution is also an important parameter to consider. For example, despite different linkages, meta isomer 1-Ph-F and ortho isomer 4-Ph-F
possess an almost identical absorption spectrum (Figure 3-Left), showing that the bridge can cancelled the effect of the linkages on the conjugation length. This is a different behaviour than that highlighted for the couple 1-Ph-SBF/4-Ph-SBF and reveals the key role played by the bridge.
As for the absorption properties, the fluorescence properties of the four SBF isomers are also very different depending on the substitution pattern and allow to obtain efficient emitters in different ranges of colour (Figure 2-Bottom, right). 4-Ph-SBF
even appears as a remarkable example discussed in detail in
Part 3. As 2-Ph-SBF and 3-Ph-SBF possess similar emission
spectra (λmax=334 and 332 nm resp.) and quantum yields in
solution (0.87 and 0.74 resp.), from a spectral shape point of view, para and meta linkages are almost indistinguishable in fluorescence. As in absorption, the emission spectrum of 1-Ph-SBF is very similar to that of its building block 1-Ph-SBF, showing
that the electronic effect of the pendant phenyl ring is also
Accepted
450 500 550 0.0 0.5 1.0 1.5 ET SBF > ET 1-Ph-SBF > ET 3-Ph-SBF > ET 4-Ph-SBF > ET 2-Ph-SBF SBF 1-Ph-SBF 2-Ph-SBF 3-Ph-SBF 4-Ph-SBF N orm a li z e d P L (a . u.) Wavelength (nm) 431438433445 484
almost erased at the excited state. To sum up, for all the isomers, the absorption and emission spectra at room temperature follow the same trend, which is driven by the nature of the linkage (ortho, meta and para) and the steric hindrance induced by these linkages. At 77K, the phosphorescence properties were found to be different as well as the role played by the discussed parameters.
As the main interest of SBF isomers is their capacity to host phosphors in PhOLEDs, their triplet state energies (ET) are
relevant data. When designing host materials for PhOLEDs, it is essential to know how the ET varies when adding molecular
fragments (Figure 4).
Figure 4. Effect of the phenyl substitution on the ET of SBF (Top) and biphenyl (Bottom)
Figure 5. 1-Ph-SBF, 2-Ph-SBF, 3-Ph-SBF and 4-Ph-SBF. Emission spectroscopy at 77 K in 2-Me-THF (Left) and SDD triplet with isovalues of 0.004 (Right), SBF is added for comparison purpose)
Thanks to the emission spectra at 77K, the ET of 1-Ph-SBF,
2-Ph-SBF, 3-Ph-SBF and 4-Ph-SBF were respectively estimated at
ca 2.86, 2.56, 2.83 and 2.78 eV (Figure 5-Left). Due to the π-conjugation disruption, the meta-substituted terphenyl core of
1-Ph-SBF and 3-Ph-SBF leads to a high ET compared to the
para-substituted terphenyl core of 2-Ph-SBF and, to a lesser
extent, to the ortho-substituted terphenyl core of 4-Ph-SBF.
Thus, the ET of 1-Ph-SBF (2.86 eV) is almost identical to that of
SBF (ET=2.88 eV) and there is no electronic influence of the
pendant phenyl. Oppositely to the conclusions drawn above for S1, the authors reported that the emission from T1 state
follows the classical para/ortho/meta sequence as the ET
increases as follows 2-/4-/3-/1-Ph-SBF. The linkage fully drives
the ET whereas the bridge and the steric hindrance are
dominant parameters at room temperature. The same effect is also reported for the fluorene series presented Figure 3, highlighting that this trend is general.39 Our group has tried to
rationalize this feature. Indeed, thanks to theoretical calculations, we have shown that the triplet exciton of 3-Ph-SBF and 1-Ph-3-Ph-SBF is exclusively localized along the substituted
fluorene, the pendant phenyl having no contribution (Figure 5-Right). Both molecules display hence a high and similar ET. In
4-Ph-SBF, as the triplet exciton is partially delocalized on the
pendant phenyl, the ET is decreased. A point noted by the
authors is that the delocalization of the triplet exciton is different to that of the HOMO and LUMO (Figure 6) highlighting a different contribution of the pendant phenyl ring. This is discussed below.
Electrochemical studies of these SBF isomers have shown that the contributions of the phenyl ring in the electronic distribution of the HOMO/LUMO are different depending on the substitution pattern (Figure 6). This is discussed below. Despite their different phenyl substitution positions, the HOMO of 1-Ph-SBF, 3-Ph-SBF and 4-Ph-SBF (ca -5.94/-5.95 eV,
Table 1) possess the same energy as that of unsubstituted SBF
(-5.95 eV) with no or weak electronic density on the pendant phenyl ring (thanks to its para linkage and small dihedral angle, the HOMO of 2-Ph-SBF lies at a higher energy, -5.86 eV). The
trend is different for the LUMO energy levels. If the phenyl ring at C1 does not influence the LUMO energy of 1-Ph-SBF (-1.73
eV, almost identical to that of SBF, -1.74 eV), it has a
non-negligible influence when located at C3 or at C4 with deeper LUMO reported (-1.77 eV and -1.87 eV resp). This is particularly pronounced for 4-Ph-SBF, which presents a
significant contribution of the phenyl ring in the LUMO distribution, contrary to its HOMO level (Figure 6-Top). Thus, the phenyl ring has a different influence on the benzenoidal HOMO/quinoidal LUMO distribution depending on the regioisomer involved. We believe that the torsion between the SBF and the phenyl ring is responsible of the different trend between HOMO and LUMO energy levels. Another series of C1-linked SBF materials have later confirmed this feature (see part 5).6 1-Ph-SBF 3-Ph-SBF 4-Ph-SBF 2-Ph-SBF 4-Ph-SBF 2-Ph-SBF ET= 2.78 eV ET= 2.56 eV Ortho-terphenyl linkage Para-terphenyl linkage ET= 2.88 eV + + 9,9'-SBF + + Meta-terphenyl linkage Meta-terphenyl linkage ET= 2.83 eV ET= 2.86 eV 1 2 3 4 1-Ph-SBF 3-Ph-SBF ET= 2.55 eV ET= 2.67 eV ET= 2.85 eV Para-terphenyl Ortho-terphenyl + + Biphenyl Meta-terphenyl ET= 2.82 eV +
Accepted
Manuscript
-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 SBF 1-Ph-SBF 2-Ph-SBF 3-Ph-SBF 4-Ph-SBF I ( n o rm al iz ed a t E 1 red ) E(V) vs SCE 1.4 1.5 1.6 1.7 -0.5 0.0 0.5 1.0 I ( n o rm al iz ed a t E 1 ox ) E (V) vs SCE
Figure 6. 1-Ph-SBF, 2-Ph-SBF, 3-Ph-SBF and 4-Ph-SBF. Top. LUMO (Left) and HOMO (Right); Bottom.CV in DMF/Bu4NPF6 0.1 M (reduction, left) and in dichloromethane/Bu4NPF6 0.2 M (oxidation, right), 100 mV/s (SBF is provided for comparison purpose).
This structure-property relationship study of the four phenyl-substituted SBFs describes well the general behaviour of all the other SBF regioisomers reported to date in the literature. Some examples of these families are described below. This study also shows the key role played by the bridge, the nature of the linkage and the fluorene/substituent angle on the electronic properties of SBF positional isomers. Such a study lays the foundation of SBF regioisomerism and provides interesting information for further materials design.
Figure 8. Molecular structures of 4-PhCbz-SBF, 4-4Py-SBF, 4-Ph-SBF, 4-Ph(Me2)-SBF, 4-Ph(OMe)3-SBF and 4,4’’-(SBF)2 obtained by X-Ray diffraction on single crystals (Front view). The substituent attached on the SBF core is (from left to right): 4-phenylcarbazole; 4-pyridine; phenyl; 3,5-dimethylphenyl; 3,4,5-trimethoxyphenyl and 4-spirobifluorene
Part 3. 4-substituted Spirobifluorenes
Figure 7. Selected examples of 4-substituted SBFs.
4-substituted SBFs are the 2nd generation of substituted SBF regioisomers, far less developed than their 2-substituted counterparts. Selected examples of 4-substituted SBFs are reported in Figure 7. It should be noted that an exhaustive review on 4-substituted SBFs was published by our group in 2017.3 The substitution at C4 of the SBF core was first
described in 2009 by the group of Ma with a C4-linked SBF dimer, 4,4’’-(SBF)2 (Figure 7).31 Our group has since intensively
studied the impact of this substitution on the electronic properties of the SBF core.32, 34, 35, 63, 64 The common feature
between all these molecules is their high ET, which has allowed
their use as efficient host materials for green and blue PhOLEDs.3 From a more fundamental point of view, many
interesting structural and electronic features have been highlighted over the years for this family of molecules. At the beginning of these studies, in 2009, the main questions to answer was: What are the consequences of the C4 substitution
on the structural and electronic properties of the SBF core and are these semi-conductors suitable for designing PhOLED hosts? Ten years of research later, we present herein some
answers.
As above-mentioned and exemplified in Part 2 with the molecule Ph-SBF, the π-conjugation breaking observed in
4-substituted SBFs is induced by the large dihedral angle between the fluorene and its pendant substituent. This allows to strongly disrupt the π-electron distribution, providing notably a high ET.
Despite more and more investigation in the last five years, the number of 4-substituted SBFs reported to date is rather limited.3 At C4 of the SBF core have been introduced (i) electron withdrawing fragments such as heterocycles (pyridine, 4-4Py-SBF/4-3Py-SBF,32 pyrimidine, 4-5Pm-SBF34) or
phosphine oxide unit (4-POPh2-SBF),65 or (ii) electron donating
fragments such as phenylcarbazole (4-PhCBz-SBF),
trimethoxyphenyl (4-Ph(OMe)3-SBF)66 dibenzothiophene (
4-DBT-SBF)67 or dibenzofuran (4-DBF-SBF), Figure 7.68 As a
function of the steric congestion induced by the substituent, the dihedral angle between the fluorene and the substituent can be drastically modified, which in turn alters the resulting
4-5Pm-SBF 4-POPh2-SBF 4-Ph-SBF R : N N POPh2 R 4-substituted SBF 4 X OMe OMe OMe N 4,4''-(SBF)2 4-DBF-SBF 4-PhCBz-SBF 4-PhOMe3-SBF OC16H33 OC16H33 OC16H33 O N H OC16H33 OC16H33 OC16H33 4-EPHDB-SBF 4-EPHDBA-SBF 4-4Py-SBF N N 4-3Py-SBF x=O x=S 4-DBT-SBF N N 4-PPI-SBF 42.2° 88.3° 51.2° 73.2°
4-PhCbz-SBF 4-4,Py-SBF 4-Ph-SBF 4-Ph(Me)2-SBF 4-Ph(OMe)3-SBF 4,4’’-(SBF)2
45.4° 78°
Accepted
electronic properties. Six selected examples of X-Ray structures are presented in Figure 8. For an unsubstituted phenyl ring (in 4-Ph-SBF) or a phenyl ring analogue such as
pyridine (in 4-4Py-SBF), the dihedral angle remains low,
around 40/50°. If a substituent is attached in para position of the phenyl ring such as in 4-PhCbz-SBF, the dihedral angle
remains in the same range. This clearly indicates that the substitution in para position does not add any significant steric hindrance. However, if a substituent is added in meta position of the phenyl ring, the dihedral angle is impressively increased (78° in 4-Ph(OMe)3-SBF and 73.2° in 4-Ph(Me2)-SBF). This
angle becomes closer to 90° when the phenyl attached at C4 is substituted in ortho position. Thus, 4,4’’-(SBF)2, which is built
on two C4-linked SBF fragments each substituted at C4 displays an angle as high as 88.3° (Figure 8-Right).31 Therefore,
the size and the substitution pattern of the substituent attached at C4 both have a strong impact on the dihedral angle, which drives the electronic properties (see below). Thanks to this large fluorene/substituent dihedral angle, there is a strong π-conjugation disruption in all the 4-substituted SBFs reported in the literature.3 In the absorption spectra of
these molecules, the intensity of the tail at around 320 nm characterizes the different degree of electronic delocalization between the fluorene and its substituent (Figure 9- Top, left). Thus, depending of the substituent borne by the fluorene (pyrimidine, pyridine, trimethoxyphenyl etc), the intensity of this tail is different and function of the angle formed between the substituent and the fluorene. This π−conjugation disruption has a direct influence on the ET values, which are for
all these molecules (i) above 2.7 eV and (ii) higher than their 2-substituted counterparts (Figure 9-Bottom, left). However, the influence of the substituent is different for each molecule leading to different ET values. The highest ET is reported for the
molecules possessing the largest dihedral angle in the series.
4-Ph(OMe)3-SBF and 4,4’’-(SBF)2 possess indeed a very high ET
of 2.84 eV and 2.81 eV, close to that of unsubstituted SBF
(2.88 eV), and respectively display the largest dihedral angle of 78.5° and 88.3° (Figure 8).34, 35 Thus, controlling the fluorene/C4 substituent dihedral angle can be an interesting strategy to control the intensity of the electronic coupling in 4-substituted SBFs. 290 300 310 320 330 340 350 0.0 0.5 1.0 4-substituted SBFs A bs or ba nc e (nor m al iz ed ) Wavelength (nm) 4-Ph-SBF 4-4Py-SBF 4-5Pm-SBF 4-Ph(OMe)3-SBF 2-Ph-SBF 2-4Py-SBF 2-5Pm-SBF 2-substituted SBFs 350 400 450 500 0.0 0.2 0.4 0.6 0.8 1.0 N or m al iz ed P L ( a. u. ) Wavelength (nm) 2-substituted SBFs 4-substituted SBFs 440 460 480 500 520 540 560 0.2 0.4 0.6 0.8 1.0 Wavelength (nm) N or m al iz ed P L I nt ens it y ( a. u. ) N N N N MeO MeO OMe N N 4-5Pm-SBF ET= 2.75 eV 4-Ph-SBF ET= 2.77 eV 4-4Py-SBF ET= 2.74 eV 2-5Pm-SBF ET= 2.58 eV 2-Ph-SBF ET= 2.56 eV 2-4Py-SBF ET= 2.58 eV 4-Ph(OMe)3-SBF ET= 2.84 eV
Figure 9. Top. UV-vis absorption (normalized at 309 nm, left) and emission spectra at room temperature (normalized at λmax, right) in cyclohexane, of selected examples of 2- and 4-substituted SBFs. Bottom. Emission spectra at 77K in 2-Me-THF (Left) and molecular structures of reported compounds (Right).
The most intriguing particularity of 4-substituted SBFs is their uncommon fluorescence. At room temperature, almost all the 4-substituted SBFs reported exhibit similar fluorescence spectra. These spectra are large, structureless (350/380 nm) and strongly red-shifted compared to the other positional isomers substituted at C1, C3 and especially at C2, which display resolved emission bands (Figure 9-Top, right). It is indeed known for C2-isomers that the C-C bond linking the pendant substituent and the fluorenyl core displays a double-bond character in the excited state, rigidifying the structure and leading to a resolved spectrum.69, 70 Thus, 4-substituted
SBFs display a fluorescence emission in a different range than that of the other positional isomers. Why do these isomers
possess such very different emission spectra? Despite no
complete answer having been provided yet,32, 34, 35, 64 some
findings recently reported by our group are presented below as a representative example with the couple 4-PhCbz-SBF/4-Ph(OMe)3-SBF.33 The fluorescence spectrum of 4-Ph(OMe)3
-SBF is characteristic of a 4-substituted -SBF30, 31, 35, 36, 65, 67, 68:
structureless, large and presenting a significant Stokes shift. Thanks to theoretical calculations, the origin of this large Stokes shift and unusual large fluorescence has been explained by the significant difference between the geometries of the ground (S0) and first singlet excited (S1) states of 4-Ph(OMe)3
-SBF (Figure 10-Right). Thus, due to the particular molecular
arrangement of the 4-substituted SBF scaffold, in which the pendant substituent is sterically hindered by the hydrogen atoms of the substituted fluorene,35 the observed emission
results from a large distribution of conformers. On the contrary, 4-PhCbz-SBF presents a well resolved emission
spectrum and a small Stokes Shift (Figure 10-Left). As can be seen in Figure 10-Right, the geometries of S0 and S1 are indeed
very similar translating very weak molecular rearrangements between the two states. This behaviour is most likely induced by the bulkiness of the phenylcarbazole unit attached at C4.
Figure 10. Left. Normalized emission spectra (cyclohexane) of 4-PhCbz-SBF (λexc=295 nm, orange) and 4-Ph(OMe)3-SBF (blue, λexc=309 nm), Right. Superimposition of the S0 (ground state) and S1 (first singlet excited state) molecular structures obtained by molecular modelling of 4-Ph(OMe)3-SBF (S0: sky blue, S1: green) and 4-PhCbz-SBF (S0: pink, S1: orange).
In 2009, at the very beginning of the investigation on 4-substituted SBFs, there was one important question to address: What is the effect of the SBF substitution pattern on
the HOMO/LUMO energy levels? Indeed, the introduction of
electron-donating or electron-withdrawing functional groups to adjust the HOMO/LUMO energy levels of a molecule is a
4-Ph(O 350 400 450 500 0.0 0.2 0.4 0.6 0.8 1.0 N o rmalized F lu o rescen ce (a. u .) λ (nm)
Accepted
Manuscript
10 | J. Name., 2012, 00, 1-3
widely used strategy in the field of organic electronics. However, the goal was to find out about the magnitude of the HOMO/LUMO tuning and notably the difference compared to the first generation of 2-substituted SBFs. This is an important step forward in the design of functional materials. Thus, the literature shows that the introduction of electron-donating or electron-withdrawing groups at C4 allows to tune the HOMO and LUMO to a lesser extent than at C2. Due to the partial π−conjugation disruption at C4 of SBF, the reported
electrochemical data indicate that 4-5Pm-SBF34 and 4-4Py-SBF
display a higher LUMO energy level than their corresponding 2-substituted isomers 2-4Py-SBF and 2-5Pm-SBF.34 The SBF
core is therefore less influenced by the electronic effects at C4 than at C2. Nevertheless, the electronic properties of 4-substituted SBFs can be drastically changed as a function of (i) the dihedral angle between the substituent and the fluorene and (ii) the nature of this substituent itself (length of its π-conjugated system / bulkiness etc).71 This characteristic should be considered when designing functional materials based on the 4-substituted SBF scaffold.
As mentioned in the introduction, the new generations of SBFs have been designed for organic electronics and especially to host blue and green phosphors in PhOLEDs. Their high ET and
very good thermal/morphological stabilities are key properties which have allowed 4-substituted SBFs to reach in some cases good performance when incorporated in devices (Table 3).3
For example, we can cite 4-Ph(OMe)3-SBF33 for which the EQE
are reported at ca 10% in blue PhOLEDs and at ca 20% in green PhOLEDs, 4-PPI-SBF for which the EQE is reported at ca 17% in
green PhOLEDs.72 and 4-DBF-SBF67 for which the EQE is
recorded at ca 11% in blue PhOLEDs. Incorporation of an electron accepting phosphine oxide at C4 (4-POPh2-SBF) has
led to even higher performance, displaying an EQE of 17.2% in blue PhOLEDs.65 This molecule has even been incorporated in
single-layer PhOLEDs, with Ir(ppy)3 as the emitter, leading to
an EQE of ca 13%.63 Some of these molecules have also been
tested as hosts in white PhOLED such as 4-DBT-SBF, which
displays an interesting EQE of 16.9%.67 However, the next
generations of C1 and C3-linked SBFs have in fact outperformed C4-linked SBFs in hosting phosphors in PhOLEDs (see below).
Other applications for 4-substituted SBFs have also started to emerge in the literature. We can cite for example the luminescent liquid crystalline phases generated from
4-EPHDBA-SBF and 4-EPHDB-SBF (Figure 7)64 or the association
of the 4-SBF scaffold with a diketopyrrolopyrrole fragment to construct electron donors for solar cells.73
Part 4. 3-substituted Spirobifluorenes
Figure 11. 3-substituted SBFs reported to date.
In the 3rd generation of SBF positional isomers, the
π−conjugation disruption is not directly linked to steric effects but arises from electronic effects induced by the meta linkage. The first 3-substituted SBFs have been reported in 2013 by the group of Liao and Jiang.45 In order to reach high-performance PhOLEDs, this group has particularly developed 3-substituted SBFs incorporating functional groups such as phosphine oxide (3-POPh2-SBF and 3-POPh-(SBF)2),45 triazine (
3-PhTriaZCbz-SBF),74 or carbazole oligomers (3-(3-PhCbz)-SBF,
3-(N-Ph-Cbz)-SBF and 3-diNCbz-3-(N-Ph-Cbz)-SBF).38, 75 However, until now, less than 10
examples have been reported in the literature making the molecular diversity of 3-substituted SBFs rather poor (Figure 11). In the light of the PhOLEDs performance using 3-substituted SBFs as host, among the highest reported to date for all the colours (see below), it is clear that this family will be much more developed in the coming years.
4,4''-(SBF)2 3,3''-(SBF)2 3,4''-(SBF)2 300 350 0.0 0.2 0.4 0.6 0.8 1.0 Wavelength (nm) 3,3’’-(SBF)2 3,4’’-(SBF)2 4,4’’-(SBF)2 N orm al iz ed A bs (a .u) 400 450 500 550 600 650 0.0 0.2 0.4 0.6 0.8 1.0 Wavelength (nm) P hos phore sc enc e (a .u) ET >ET>ET
Figure 12. 3,3’’-(SBF)2, 3, 4’’-(SBF)2 and 4,4’’-(SBF)2. Top. Molecular structures. Bottom. UV-vis absorption spectra (Left). Emission spectra at 77 K (Right)
As exposed in Part 2 for the phenyl isomers 3-Ph-SBF and 4-Ph-SBF, the conjugation between the phenyl and the fluorene is
more intense in the former than in the latter showing that the steric effect is dominant over the electronic effect to disrupt the conjugation. Is this true for all the 3-substituted SBFs
reported in literature? It would seem so. Indeed, the SBF
dimers 3,3’’-(SBF)2,3,4’’-(SBF)236 and 4,4’’-(SBF)231 reported by
Jiang and Liao’s group (Figure 12-Top) 36 only differ by the
substitution pattern of the two SBF cores (meta,
meta-ortho, and ortho-ortho linkage respectively) and offer herein
relevant model compounds. The absorption spectrum of
3,3’’-(SBF)2 displays its highest absorption band at 328 nm, this
band being absent for 3,4’’-(SBF)2 and 4,4’’-(SBF)2,31Figure
22-Bottom, left). Thus, the electronic coupling between the two SBF backbones seems to be more efficient in 3,3’’-(SBF)2 than
in 3,4’’-(SBF)2 and 4,4’’-(SBF)2. This trend is the same as that
exposed in part 2 for phenyl-substituted SBFs and can
3-POPh2-SBF 3-Ph-SBF : P R 3-substituted SBF 3 3,4''-(SBF)2 R N N POPh2 N N O 3,3''-(SBF)2 3-POPh-(SBF)2 3-diNCbz-SBF N N N N 3-(N-Ph-Cbz)-SBF
Accepted
Manuscript
consequently be correlated to the fluorene/fluorene dihedral angle which decreases from 88° for 4,4’’-(SBF)2 to 62° for
3,4’’-(SBF)2 and to 38° for 3,3’’-(SBF)2 (note that these values have
been obtained from X-Ray for 4,4’’-(SBF)2 and from molecular
modelling for 3,4’’-(SBF)2 and 3,3’’-(SBF)2). This clearly shows
that the steric effect is dominant over the electronic effect and therefore drives the absorption properties. The same trend is observed for the ET: 3,3’’-(SBF)2 (2.68 eV) < 3,4’’-(SBF)2 (2.76
eV) < 4,4’’-(SBF)2 (2.81 eV), Figure 12-Bottom, right, which is
this time different than that reported for 3-Ph-SBF and 4-Ph-SBF (ET of 3-Ph-SBF is higher than that of 4-Ph-SBF, Part 2). At
this stage, these two trends remain difficult to rationalize and deserve to be carefully investigated in the future. Thus, in the SBF family, the meta linkage allows a π-conjugation extension, which is different compared to other π−systems such as terphenyls.58, 62 In the light of the literature, the same
conclusions regarding the π-conjugation disruption can be drawn for most of the other 3-substituted SBFs reported to date such as 3-Ph-TriaZCbz-SBF74 or 3-diNCbz-SBF.38 The
conjugation is disrupted but not completely broken. Thus, 3-diNCbz-SBF and its 2-substituted derivative possess a similar
energy gap (the electronic coupling is nevertheless more intense for the latter) but a different ET value, the former being
higher than the latter (ET=2.82 eV and 2.72 eV respectively).38
In this example, one can hence again conclude that the ET is
driven by the linkage (meta vs para).
From the examples of 3-substituted SBFs reported to date, one can conclude that an electronic coupling between the fluorene and the attached substituent exists, although weak. This is an important feature for the future design of SBF based materials and more fundamentally when thinking about the impact of a
meta linkage on the electronic properties.
As only a few comparative studies have been reported, it is difficult to draw a general trend on the impact of a substituent on the HOMO and LUMO energy levels. However, as 3-diNCbz-SBF displays a slightly lower HOMO energy level than its
2-substituted isomer (-5.63 eV vs -5.57 eV), it seems that, as observed for the 4-substituted SBFs above exposed, the impact of the substituent on the HOMO/LUMO energy levels is weaker for a 3-substituted SBF than for a 2-substituted SBF.38
Similarly, the LUMO energy level of 3-POPh2-SBF (-2.56 eV) is
higher than that of its 2 substituted analogue 2-POPh2-SBF
(-2.65 eV).45 However, only few data exists today and it appears
premature to draw a precise structure-properties relationship map. This point will deserve to be carefully investigated in the future.
To conclude, it is important to mention that very high performance, for green and blue PhOLEDs, was reached with 3-substituted SBFs as hosts, making this family of organic semi-conductors very appealing. The first examples of incorporation in a PhOLED of a C3-linked SBF, 3-POPh2-SBF and
3-POPh-(SBF)2,45 had shown the potential of this platform. Indeed,
when hosting FIr6 in a PhOLED, EQE of 13.6% and 10.2 % were respectively obtained for 3-POPh2-SBF and 3-POPh-(SBF)2).45
These EQE values were relatively high and promising for a first example. This work revealed that the C3 position of SBF could solve the problem of low triplet energy using the traditional C2 position and has opened new paths in the field of high ET host
materials. Used as a host in blue PhOLEDs or in green TADF OLEDs, 3,4’’-(SBF)2 has displayed very high performance (EQE
of ca 22% for both). This value, which has now been exceeded by C1-linked SBF dimers (see part 5), was the highest reported for a blue PhOLED using a PHC material in 2015, showing not only the efficiency of this platform but also its versatility. Similarly, 3-(3-PhCbz)-SBF, 3-(N-Ph-Cbz)SBF75 and
3-diNCbz-SBF38 have also displayed high EQEs of ca 18/19% in blue
PhOLEDs. White PhOLEDs have also been constructed with these molecules as hosts with EQE exceeding 25 % in the case of 3,4’’-(SBF)2 and 40% in the case of 3-diNCbz-SBF.
Thus, the 3-substituted SBF scaffold has appeared in the last years as a very promising building unit to construct high efficiency host materials for PhOLEDs and TADF OLEDs. Compared to the C4-linked SBFs, the electronic properties of C3-linked SBFs can be more easily controlled and their performance in electronic devices are higher. The latest generation of C1-linked SBFs described below is even better.
Part 5. 1-substituted Spirobifluorenes
Figure 13. 1-substituted SBFs reported to date
1-substituted SBFs are the latest generation of SBFs. To the best of our knowledge, only six examples of 1-aryl-substituted SBFs for organic electronics have been reported (Figure 13).6, 76
It must be noted that a SBF possessing at C1 a methyl group ( 1-Me-SBF) has also been reported but as a model compound to
study exciton splitting.77
The aryl substitution at C1 combines the two previously described advantages of the substitution at C3 (meta linkage) and at C4 (strong steric hindrance). This combination has appeared to be the most efficient to break the π-conjugation and hence keep a very high ET (see part 2). This breaking has
been mainly assigned to the high dihedral angle formed between the substituent and the fluorene (the electronic decoupling linked to the meta linkage does not lead to a complete π-conjugation breaking) and is therefore caused by a steric parameter. This particularity has been advantageously used to design high ET host materials for PhOLEDs (see below
with the C1-linked SBF dimers). The singular geometry of the 1-substituted SBF scaffold also leads to another appealing characteristic, which should be used in the future years: the cofacial arrangement between the substituent and the facing
1-Ph-SBF : R 1-substituted SBF 1 N R 1-Cbz-SBF 1 2 3 4 Me 1-Me-SBF 1,4''-(SBF)2 1,2''-(SBF)2 1,3''-(SBF)2 1,1''-(SBF)2
Accepted
Manuscript
0.5 1.0 1.5 2.0 2.5 -10 0 10 20 30 I(µ A ) E(V) vs SCE
fluorene. Indeed, when a substituent is linked at C1 of a SBF core, preliminary studies have shown76 that there are through space interactions between this substituent and its facing fluorene (see molecular arrangement in Figure 14). Interestingly, these interactions have strong consequences on some electronic characteristics (HOMO energy level) while keeping others unaltered (ET). Only two examples have
pointed out this electronic characteristic to date. 6, 76 We
believe that others will come soon.
Figure 14: 1-Cbz-SBF. Left. Emission spectrum at 77 K (2-Me-THF, λexc = 300 nm), inset. SDD (isovalue= 0.004). Right. CV (100 mV s-1 , CH
2Cl2/Bu4NPF6, 0.2 M).
Thus, it has been shown that the first oxidation wave of 1-Cbz-SBF is strongly shifted towards lower potentials (1.17 V vs SCE,
Figure 14-Right) compared to its structurally related analogue fluorene-1-carbazole (1.30 V vs SCE).[72] This electron transfer
in 1-Cbz-SBF has been assigned to the oxidation of a cofacial
fluorene/carbazole dimer and is the consequence of strong through-space interactions between the fluorene and the carbazole. This work has also shown that the ET of 1-Cbz-SBF is
not deeply influenced by this interaction since the ET is kept
high, 2.84 eV (Figure 14-Left) very close to that of unsubstituted SBF (2.88 eV). This high ET has been explained by
the localization of the triplet exciton, which is exclusively spread out on the substituted fluorene with no contribution of the pendant substituent (Figure 14-Left, inset). Therefore, these studies have shown that the position C1 is ideal to keep a very high ET and can increase in the meantime the HOMO
energy level through π−π interactions. This particularity does not exist for the other isomers and appears as an interesting tool, which will be surely used to design host materials for PhOLEDs or more generally other functional materials for electronics.
The last examples we wish to conclude with have shown the high potential of the C1-SBF scaffold. Indeed, in early 2019, our group and that of Jiang have designed highly twisted SBF dimers linked from the C1 position : 1,1’’-(SBF)2, 1,2’’-(SBF)2,
1,3’’-(SBF)2 and 1,4’’-(SBF)2, Figure 15. These dimers have
displayed the highest performance ever reported for PHC materials (EQE of ~23% for 1,3’’-(SBF)2) when used as host in
blue PhOLEDs.6
These C1-linked dimers represent a similar isomers series to the phenyl-SBFs mentioned in section 2, with nevertheless striking differences. Indeed, the strong steric congestion imposed by the two SBF units leads to a different trend to that exposed for phenyl-SBFs. In addition to the nature of the linkage (ortho, meta and para) between two fluorene units, the dihedral angle between them is of chief importance in regards to the electronic coupling/decoupling.39 This dihedral
angle increases as follows: 54.9° for 1,2’’-(SBF)2, 57.9° for
1,3’’-(SBF)2, 61.1° for 1,1’’-(SBF)2 and 76.9° for 1,4’’-(SBF)2 (Figure
15-Bottom, right). In the case of both 1,2’’-(SBF)2 and
1,3’’-(SBF)2, the authors noted that this angle is impressively larger
than those reported for a non-encumbered phenyl/fluorene linkage,36, 39 such as in 2-Ph-SBF and in 3-Ph-SBF (Figure 2).39 In
1,4’’-(SBF)2, the presence of one SBF in ortho position of the
other leads to an impressive increasing of the dihedral angle, recorded at 76.9°. The case of 1,1’’-(SBF)2 was more surprising
since its dihedral angle (61.1°), despite the strong steric congestion imposed by the two linked C1 positions was decreased compared to less encumbered 1-Ph-SBF (75.4°,
Figure 2).39 The explanation provided by the authors is that the
angle is lowered in 1,1’’-(SBF)2 in order to minimize
the π−π interactions between the two sets of cofacial fluorenes. Thus, in the four C1-linked dimers, the relative position of the SBF fragments provides different molecular arrangements with specific steric hindrances at the origin of their electronic properties.
Figure 15. 1,1’’-(SBF)2, 1,2’’-(SBF)2, 1,3’’-(SBF)2 and 1,4’’-(SBF)2. Top. Absorption at room temperature in cyclohexane (Left) and emission at 77 K in 2-Me-THF (Right, λexc=280 nm). Bottom. Left. CV (100 mV/s, CH2Cl2/[Bu4NPF6] 0.2 M). Right. Crystal structures.
From the CVs (Figure 15-Bottom, left), the HOMO energies reported for 1,2’’-(SBF)2, 1,3’’-(SBF)2 and 1,4’’-(SBF)2 were almost identical,
-5.95, -5.95 and -5.92 eV respectively, Table 2. Thus, despite their different substitution patterns, the three dimers display very comparable HOMO energy levels, similar to that of their building block SBF (-5.95 eV). Again, this is a different behaviour than that
highlighted in the phenyl series (Table 1, the HOMO of 2-Ph-SBF
was the highest in the series), indicating the importance of the steric hindrance on the HOMO energy levels. Thus, due to the large
350 400 450 500 550 0 2x106 4x106 6x106 8x106 1x107 PL I nt ens it y ( a. u. ) Wavelength (nm) τ= 5.0 s ET = 2.84 eV 438 nm 0.0 0.5 1.0 1.5 2.0 2.5 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 I nor m al iz ed (a .u. ) E(V) vs SCE 400 420 440 460 480 500 520 540 0.0 0.5 1.0 1.5 Nor malized PL ( a. u. ) Wavelength (nm) ET 1,3'' > ET 1,4'' > ET 1,1'' > ET 1,2'' 290 300 310 320 330 340 0 1x104 2x104 3x104 4x104 E ps ilon ( mol. L -1.cm -1) Wavelength (nm)
